Abstract Cancer cell populations are marked by significant intratumoral heterogeneity and this heterogeneity varies over time with disease progression and with response to therapies. To monitor heterogeneity among large cell populations, barcode sequencing is useful for quantifying relative lineage frequencies with high resolution. However, current approaches preclude the simultaneous isolation of a particular lineage of interest. In the study of cancer and the development of clinical treatments, there are many instances in which it would be useful to perform subsequent molecular characterization or cellular functional assays on purified populations of cells exclusively of one specific lineage. A further challenge is that it is often not possible to identify which lineage is of interest until a longitudinal study has been completed?for example, comparing the survival and relative fitness of many cell lineages as they proliferate over many cell generations in a tumor. Only at the conclusion of a lineage tracing study does it become clear which specific lineages had a survival advantage. In this proposal we develop a novel lineage tracing method, Barcode Assisted Ancestral Recall (BAAR), that allows for high- resolution lineage tracking and subsequent isolation of purified cell lineages for downstream analysis. Lineage tracing via barcoding is typically a destructive measurement; with the BAAR system we gain the ability to return to an earlier time point in the evolutionary trajectory and retrieve selected cells of interest. The ability to concurrently track clonal fitness dynamics and generate lineage specific genomic and transcriptomic data over longitudinal studies will give us unprecedented insight into the behaviors of heterogeneous populations of cancer cells.